Fall 2003

IMA Workshop 1:

Statistical Methods for Gene Expression: Microarrays and Proteomics

September 29-October 3, 2003

Organizers:

Michael A. Newton
Department of Statistics and Department of Biostatistics and Medical Informatics
University of Wisconsin-Madison
newton@stat.wisc.edu
http://www.stat.wisc.edu/~newton/

and

Giovanni Parmigiani
The Sidney Kimmel Comprehensive Cancer Center and Department of Biostatistics
Johns Hopkins University
gp@jhu.edu
http://astor.som.jhmi.edu/~gp

Technological advances and resources created by genome sequencing projects have enabled biomedical scientists to measure precisely and simultaneously the abundance of thousands of molecular targets in living systems. The effect has been dramatic, not only for biology, where now the cellular role for all genes may be investigated, or for medicine, where new drug targets may be found and new approaches discovered for characterizing and treating complex diseases; the effect has also been dramatic for the mathematical and computational sciences. To assess a major component of this activity, the IMA workshop will focus on statistical issues in the study of gene expression both at the level of RNA transcripts and proteomics. Leading investigators will describe current research and future challenges in this area. The workshop format entails three lectures per day and extensive opportunity for fruitful discussion and interaction.

Various microarray technologies for measuring RNA transcript abundance have created some challenging statistical problems. There are many sources of variation in a typical experiment and these can be accounted for using statistical design and analysis-of-variance methodology; but careful attention has to be given to the high-dimensionality and complicated interactions. Statistical methods invoked early in the data analysis pipeline can remove systematic errors and improve subsequent inferences. Robust statistical methods are important to account for influential observations that may be hidden in massive data sets. A wide range of supervised and unsupervised learning methods have been considered to better organize data, be it to infer coordinated patterns of gene expression, to discover molecular signatures of disease subtypes, or to derive various predictions. Related efforts aim to reconstruct regulatory networks from large sets of expression data. Theoretical problems arise in statistical inference when attempting to address thousands of gene-specific hypotheses at once, such as the problem to bound the rate of false detections of differential expression. Further, research in statistical computing concerns infrastructure to enable efficient and flexible calculations with large expression data sets. The workshop will consider these and other pressing problems generated in current research.

LIST OF CONFIRMED PARTICIPANTS